Long Form

Birdman wrote on 3/23/2005, 7:17 AM
Hi Folks,
I have been doing :30's and :60's with some regularity in Vegas, but now it looks like I've contracted for a 12 - 15 minute promo for a small city.

My concern is the most efficient manner in which to build and render the final product. I will be utilizing various footage, graphics, stills and limited transitions. As this will be an ongoing project for 30 - 60 days, I will be shooting, capturing to Vegas, shooting, capturing, etc.

The project will be in a "chapter" format, i.e, fade in, graphic "Quality of Life", crossfade to testimonial, voice under of testimonial with video/stills, fade out to new testimonial. Each "chapter" will be 2 - 3 minutes.

Could I complete chapter 1, render, then build chapter 2, render, etc. without having to rerender the whole video when all chapters are finished?

This question may seem somewhat convoluted, but so am I, for that matter.

Any assistance would be appreciated.

Thanks - David Bird

Comments

Spot|DSE wrote on 3/23/2005, 7:23 AM
You could build each chapter as a separate veg, and then render them as individual chapters, placing each chapter on the DVD. You can also build them all, render them in a lo/no compression format, and stitch them all together as one large file when you're completely finished. For DVD output, we do a lot of work this way, rendering to the 4:2:2 YUV codec so that our archives are high quality, but then we take the multiple YUV files and render them as a single MPEG 2 when we're done. This gives us archivable files and a DVD output.
rmack350 wrote on 3/23/2005, 7:25 AM
You could build them as seperate veg files and render each one as you go.

It depends a bit about what the final deliver format is. If they want a dvd then you could do mpeg renders (using dvd architect template). If they want tape then you could do renders to whatever format you think best to make tapes from. Probably DV but not neccessarily.

My current thinking is that if you're going to DVD it's best to avoid making a DV25 intermediate because your text and graphics will suffer.

Others with better experience will soon offer some good ideas.

Rob Mack
Randy Brown wrote on 3/23/2005, 7:59 AM
The only thing I would add is to make sure you keep the individual .VEGs because over that length of time you may (usually a "probably" for me) wind up changing something before the final render.
Randy
Birdman wrote on 3/23/2005, 8:06 AM
Thanks guys,
Ok, the way I understand it, I would complete a chapter, render, and save as an individual veg file. After I complete the chapters, I would then bring them all back onto a master time line? Would that master time line have to be rendered again? If so, since the individual veg files have been rendered, would that make the master time line render pretty quickly? (Boy, I'm gettin' tired of saying "render")
Thanks - David Bird
Spot|DSE wrote on 3/23/2005, 9:05 AM
The master timeline will need to be rendered if it's a long file to be compiled, yes. If you render the individual files to DV, then the final render to a single file will be lightning fast, because it's just a bit for bit copy. How are you planning to deliver though? Tape or DVD?
riredale wrote on 3/23/2005, 9:09 AM
Building the chapters separately is the way I've done my documentaries for the past few years. Each chapter is maybe 5-10 minutes in length and is an entirely separate veg file from the rest of the project.

Once I'm happy with the whole project I render the separate chapters as separate "intermediates." I then bring those intermediate avis onto a new timeline, adjust the timing between them, and then render a long final avi, typically about 2 hours in length. I do the same for the front and rear audio (for a surround-sound project). Those finished files then go to the MPEG2 encoder and AC-3 encoder.

I like doing it this way because it seems easier for me to have complete control over a short clip than a long one. Also, given the compexity in each chapter, the turnaround time is greatly reduced if (or I should say "when") I note in the finished intermediate that I should have held some scene longer, done a J-cut on a particular transition, increased the audio level somewhere, or whatever.

Using intermediate files means an increased need for disk space, but big deal; when you can get 120GB for $50, using 26GB (2 hour's worth) of additional disk is a non-issue.

Finally, everything I do is in the DV25 format, since that's the way it comes from the camera.
Birdman wrote on 3/23/2005, 9:18 AM
Spot,
Ahhh...delivery. To quote the client "Don't you think it would be a good idea to make some DVD's and some VHS tapes. Some people might not have a DVD player." And I'm thinking....if they don't have a DVD player, loan them 50 bucks so they can go buy one....but of course I THOUGHT it as opposed to saying it! I'mgoing to push for one or the other, not BOTH...ARGGGHH
David
Spot|DSE wrote on 3/23/2005, 9:27 AM
K...Depending on content, I'd go one of two ways.
1. IF, IF> your media is mostly DV acquired, and you don't have a lot of stills or generated media (titles, etc) then I'd render the individual chapters to DV, and use those packaged intermediates as your files to create a master. Print to tape for the VHS masters.

2. If your media is mostly stills, graphics, titles, non-DV originated, I'd render the packages to 4:2:2 YUV, it'll take longer, but will give you a slightly better MPEG encode. You'll then do a Print to Tape of that as well, for your VHS.
This is a slightly longer, but slightly higher output quality format, and it's better for archiving if you later decide to do something with the media.
Birdman wrote on 3/23/2005, 9:30 AM
Spot,
All video will be DV....I anticipate 6 to 8 minutes of the total...WITH stills, graphics, titles,etc...
db
Spot|DSE wrote on 3/23/2005, 9:38 AM
You have to amortize time vs what you're gonna do with it then. If the majority of the project is DV, you gain little by going to YUV. Only titles, transitions, and graphics will benefit from the higher quality format.
BrianStanding wrote on 3/23/2005, 9:50 AM
I don't disagree with what's been said here, but 12-15 minutes total length may not justify the multiple .VEG/ intermediate render / master .VEG approach. I usually keep everything in one .VEG file until I break the half-hour mark or more. You can use regions and subclips to help you navigate or select portions of the timeline.

Unless you're dealing with dozens of tracks, Vegas should be able to accommodate 12-15 minutes of material on a single timeline just fine. Advantages to this approach include:
- being able to quickly apply track-level effects or pan/crop to similar media types all at once;
- saving all that render time. No renders needed until you're satisfied and do a one big render at the end.
- facilitates recapture and re-editing at a later date, since you don't have to hunt through your multiple veggie files to find the right one. Remember, Vegas destroys camera-original timecode whenever it renders out a new file. This means you have to recapture the original clips then re-render a new file.

Disadvantages include:
- you have to be very careful about how edits (especially ripple edits and group/ungroup actions) affect downstream events on the timeline. I find locking all events except the segment I'm working on helps avoid nasty problems.

To each their own.
Birdman wrote on 3/23/2005, 10:00 AM
Brian,
Thanks for your input....As I initially indicated, all of my previous work has been of the 30 and 60 second commercial variety, so I'm venturing into new territory...ANY assistance is appreicated.
Thanks - David
BrianStanding wrote on 3/23/2005, 10:06 AM
Others will disagree, but I'm a big advocate of using the Trimmer to make subclips from the longer material, then organizing all your subclips into Media Bins in the Media Pool. Get all this stuff organized and readily accessible in some way that makes sense to you before you put ANYTHING on the timeline.

I've done two-hour + projects in Vegas this way, with good success.
BillyBoy wrote on 3/23/2005, 10:14 AM
You know those cages with big wheels you see some chipmunk in, running on the wheel like crazy and never getting anywhere? That's what immediately comes to mind everytime anybody starts explaining their convoluted, multiphase rendering schemes or I got to have 50 different VEG files, etc, etc, and then the icing on the cake is they're saying a "long" project is maybe 15-20 minutes. To each his own preferred method of course... but geez, why go through so much extra work?

I sometimes make REALLY big projects. A hour, even close to two hours long and having dozens, even over a hundred source files or more in the project. I always make only one veg file, regardless how complex the project is going to be and I never ever render until its finished. Never. I'm not suggesting my way is "better" just trying to understand why some insist on such elborate pre-steps, which would drive me crazy.

Vegas is non destructive. Your source files regardless how many or where they are on your system remain as-is. I just don't see all the unnecessary gymnastics, starting, stopping, making endless prerenders or intermediate renders and keeping track of dozens of seperate veg files, etc., etc.
Birdman wrote on 3/23/2005, 10:38 AM
BillyBoy,
I appreciate YOUR input, too. Hummm....within the parameters I'm used to working, 15 - 20 minutes IS A LONG PROJECT. That's why I'm asking for assistance. Not knowing the issues surrounding longer form work , it makes sense to me to ASK those who are willing to share their experience with others. Since I have never been exposed to this type of work, I welcome any assistance that is proffered. Including yours. I'll review all the suggestions, try some or all, and then find a solution that works for me. I'm just not smart (or arrogant) enough to know the very best way to do any project. And even if I did, I sure wouldn't flame someone who has taken the time to offer me their thoughts and opinions.
Regards - David Bird
Spot|DSE wrote on 3/23/2005, 10:39 AM
If you're working with analog-originated captures, or you don't use scene detect, this is the best way of working, IMO. setting up bins in advance, etc helps a lot. However, Birdman's project is only 15 minutes, not longform at all in a sense of reality.
Everyone's got their way of working, but for speed and management, if you do this on a grind basis, media management is always best in packages. I'm sure that there are those that work with single timelines that are longer than an hour, but that's hard on the proc, hard on the eyes, and definitely a challenge when rippling, inserting, etc. All the news rooms that are using Vegas that I'm aware of, for example, cut packages, then assemble the packages as a final show. This is how the Nightline crew is working on even their teasers. Not that theirs is the only way...but it's sorta the "common" way.

Nice thing about Vegas is it accomodates all workflows.
BillyBoy wrote on 3/23/2005, 10:55 AM
Seems there is a thin skinned epidemic in this forum. Very contagious apparently. You know what? Since some think I should keep any opinion I have to myself, then I will. Since nobody cares what I think, there's little point in my writing any more tutorials either. Would just be a waste of my time.


DavidMcKnight wrote on 3/23/2005, 11:00 AM
The other nice thing about Vegas is that you CAN work in hour-long timelines, if you want. I came from a system that you had to prerender anything over 10 - 12 minutes and assemble into a final project (Pinnacle Studio). otherwise it would crash in a most spectacular way.

Most Vegas projects I do are about 30 - 40 minutes and I have done it both ways. I did one almost hour long project on a single timeline and I think for anything that long I would break it up for sure. But only if I knew the sub clips weren't going to be changed.

<edit for clarity>
Birdman wrote on 3/23/2005, 11:05 AM
Thanks to everyone for their assistance...I think I have enough knowledge, now, to be dangerous! Now the fun part....gotta go shoot some footage (as soon as it stops raining)
Regards all - David Bird
dand9959 wrote on 3/23/2005, 12:43 PM
Seems to me your workflow (re: everything on one long timeline or multiple .veg files) depends also on how conducive your project is to having separate .veg files per "chapter". In a long, contiguous narrative piece, it may be the case that the only clear "breaks" or chapter possibilities occur at a specific cut between scenes. This may NOT be a good place to break the project into separate .veg files: what if you want to remove the cut, or change the dissolve. You'll have to wait to do this at the assembly stage.

On the other hand, the nature of your project may very well lend itself to working with each segment as a separate project. Weddings, sporting events, project with clearly delineated segments, etc. (It sounds like this is the case for your project.)

It seems to me that- in general - a 15-20 minute project can be handled pretty easily as one project. It's just a matter of personal preference and how good your organizational skills are. :-)

I've done it both ways (<insert your own sexually-oriented joke here>), and recognize there are pros/cons to each.
jaegersing wrote on 3/23/2005, 4:50 PM
For me, if the total duration is 15 minutes I would keep everything in one project. I'd make full use of timeline markers to keep track of the various segments or chapters, and probably I'd render specific sections as DV files as I went along, just to check the output in Best mode on a TV monitor. The idea is not to have any nasty surprises at the end, so that the "final" render is only done once (not counting changes demanded by the customer of course!).

Richard Hunter
PhilinCT wrote on 3/23/2005, 5:50 PM
Hi use both of the work flows suggested, The length of the project does not really direct me as to which I use as much as the complexity and number of streams. My last 60:00 doc was done with one time line, but the open, credits and "special graphics" where all done on seperate timelines, that way I could open the credit file each time I added a crew member or picked up another sponsor......

I even use seperate lines for :30 commercials, if the graphic load is high, that way I can focus on the final audio and video mix onthe main time line.

I also can not have enough "BINS" the more you have , the quicker you will find that missing still shot.

Phil
riredale wrote on 3/23/2005, 6:33 PM
BB may have a point. I've always done my projects as a series of little projects. Maybe because they're documentaries, and each day stands by itself, I started doing it this way.

I have done one long veg before; for example, I did a DVD of a Christmas play called "Amahl and the Night Visitors" a few months back. That veg was about an hour long, and things turned out fine.

One thing I do enjoy about the chopped-up style is that I get a series of small accomplishments along the way. Also, I can render to DV tape in sections, so if my PC explodes or the house burns down I still have completed sections. But probably the biggest benefit is that the render is very quick.
rmack350 wrote on 3/23/2005, 10:09 PM
Since the logging/scene detection topic came up, I'll add 2 cents.

Normally I get media that's been logged on the shoot and I like to make my clips match what the house system is using (M100). So I've been an advocate of short clips, scene detection, etc.

However, I just started a project that'll be about 8 minutes, has tons of raw footage of interviews with people who ramble. I'm finding that scene detection wasn't the right choice for this. In retrospect, this is definitely a case where capturing whole tapes and then breaking it into subclips would have worked nicely. I'd have a lot of named objects in the pool and it'd be much easier than finding the spot in clip 23 or maybe 24 or was it 22? where someone says something usable.

The one down side of subclips is that their heads and tails are locked so if you need more you have to go back to the trimmer and make a new subclip. It's not impossible but it's inconvenient when your sitting there wishing you could just drag out the end of the subclip you've already got.

I'm sure there are lots of ways to skin the logging cat but you've got to skin it.

Rob Mack